Interleukin-15-Deficient Mice Develop Protective Immunity to Toxoplasma gondii

Abstract

Previous studies have suggested an important role for interleukin-15 (IL-15) in resistance to and memory for Toxoplasma gondii infection. The studies presented here reveal that IL-15 is not required for infection-induced expansion of NK or CD8⁺ T cells. Furthermore, IL-15^−/− mice develop long-term protective immunity to this pathogen.

Interleukin-15 (IL-15) is a cytokine which is important in the development and homeostasis of several lymphocyte populations including NK, NKT, IEL, and CD8⁺ T cells (5, 10). In addition, IL-15 can also stimulate proliferation, enhance cytotoxicity, and upregulate production of gamma interferon (IFN-γ) by NK and T cells (3, 7). Resistance to the intracellular parasite Toxoplasma gondii is dependent on the ability of NK and T cells to produce IFN-γ, and infection with this pathogen leads to increased levels of IL-15 mRNA (4, 8). Therefore, it has been proposed that IL-15 would be an important factor in immunity to T. gondii (11, 12). This hypothesis was supported by studies in which the addition of IL-15 to splenocytes enhanced NK cell production of IFN-γ in response to parasite antigens (8). Additionally, treatment with IL-15 was shown to enhance T-cell memory responses to T. gondii (11, 12, 14).

The studies presented here used IL-15^−/− mice to address the role of IL-15 in toxoplasmosis. Intraperitoneal infection of age-matched wild-type (C57BL/6 mice; Taconic, Germantown, N.Y.) and IL-15^−/− mice (Taconic) with 20 cysts of the ME49 strain of T. gondii revealed that these mice produced similar serum levels of IFN-γ 7 days following infection (Fig. 1A). Furthermore, analysis of parasite-specific recall responses at this time point demonstrated that splenocytes from wild-type and IL-15^−/− mice produced comparable amounts of IFN-γ (Fig. 1B). Additionally, depletion of IFN-γ from infected wild-type or IL-15^−/− mice resulted in the rapid death of these mice (data not shown). Moreover, although naïve IL-15^−/− mice almost completely lack NK cells, infection with T. gondii led to a marked increase in NK cell number that was similar to that seen in wild-type mice (Fig. 1C). In addition, no obvious differences were noted in tissue histology or parasite burden at this time point (data not shown). Although wild-type and IL-15^−/− mice were both resistant to the acute phase of toxoplasmosis, the genetic background of the mice used in these studies (C57BL/6) predisposes them to develop toxoplasmic encephalitis which results in death during the chronic phase of infection. No significant differences in immunopathology or cyst burden were observed in chronically infected wild-type and IL-15^−/− mice (data not shown), and the wild-type and the IL-15^−/− mice succumbed to infection at similar rates (Fig. 1D).

FIG. 1.

IL-15^−/− mice are not deficient in early IFN-γ production or susceptible to acute toxoplasmosis. C57BL/6 mice and IL-15^−/− mice were infected intraperitoneally with 20 cysts of ME49 and samples were collected 7 dpi. (A) Serum was collected from wild-type (WT) (n = 6) and knockout (KO) (n = 5) mice 7 dpi and IFN-γ was measured by enzyme-linked immunosorbent assay. Data shown are representative of three individual experiments. (B) Antigen-specific IFN-γ production was measured by stimulating splenocytes from mice infected 7 days earlier for 48 h with soluble Toxoplasma antigen (STAg). No defect in IFN-γ production was observed. Data shown are representative of three individual experiments. (C) Splenocytes were surface stained for CD3 and NK1.1 to assess the total NK⁺ CD3⁻ cell population. Data shown are representative of two individual experiments containing at least three mice per group. (D) C57BL/6 mice (n = 18) and IL-15^−/− mice (n = 15) were infected intraperitoneally with 20 cysts of ME49 and survival was monitored. Results presented are pooled data from three independent experiments containing no less than four mice per group.

Since naïve IL-15^−/− mice have reduced numbers of CD8⁺ T cells (10), which are an important source of IFN-γ for resistance to acute and chronic toxoplasmosis, it was surprising that the production of IFN-γ in antigen-specific recall responses was not deficient in IL-15^−/− mice. Therefore, studies were performed to assess whether there were compensatory changes in the populations of infection-induced activated (CD44^hi CD62L^lo) CD4⁺ and CD8⁺ T cells in the absence of IL-15. Analysis of splenocytes from wild-type and IL-15^−/− mice (7 days postinfection [dpi]) revealed that infection led to a similar increase in the percentage of activated CD4⁺ and CD8⁺ T cells (Fig. 2A and B). However, while the spleens of infected wild-type and IL-15^−/− mice contained similar total numbers of CD4⁺ T cells (Fig. 2C), there was a marked reduction in the absolute numbers of activated CD8⁺ T cells (Fig. 2D). Nevertheless, despite these defects there is still a comparable n-fold increase (∼6-fold) in the numbers of infection-induced activated CD8⁺ T cells.

FIG. 2.

T-cell activation is not defective in IL-15^−/− mice during acute toxoplasmosis. Wild-type and IL-15^−/− mice were infected with 20 ME49 cysts intraperitoneally and splenocytes were collected 7 dpi. Splenocytes were surface stained and gated on CD4 (A) or CD8 (B), and the activation markers CD44 and CD62L were evaluated. To compare activation status with total cell number, total numbers of activated CD4⁺ CD44^hi CD62L^lo (C) or CD8⁺ CD44^hi CD62L^lo (D) T cells were plotted. This is representative of two experiments containing four mice per group. WT, wild type; KO, knockout.

Although the studies above indicate a limited function for IL-15 in resistance to a primary infection with T. gondii, previous reports have suggested that IL-15 plays a role in the maintenance of CD8⁺ T-cell memory for T. gondii (11, 13). A standard model to test memory for T. gondii infection is to vaccinate mice with the avirulent, temperature-sensitive ts4 T. gondii strain followed at least 1 month later by challenge with the virulent RH strain (6, 16). To determine if IL-15^−/− mice displayed any defects in memory responses, wild-type and IL-15^−/− mice were vaccinated once with 2 × 10⁴ ts4 parasites, followed by a challenge 8 months later with 1 × 10⁴ RH parasites. Whereas naïve mice (wild type and IL-15^−/−) succumbed to RH challenge 7 days following infection, the vaccinated wild-type and IL-15^−/− mice survived challenge (Fig. 3). In addition, mice challenged 1, 2, and 6 months following vaccination did not succumb to infection. At all time points examined, wild-type and IL-15^−/− vaccinated mice produced antigen-specific IFN-γ (data not shown). Not only was there no defect in the production of antigen-specific IFN-γ, the IL-15^−/− mice consistently produced more IFN-γ, though this increase was not statistically significant.

FIG. 3.

Immunized IL-15^−/− mice are resistant to virulent T. gondii challenge. Mice were vaccinated with the avirulent ts4 parasite strain and challenged 8 months later with the virulent RH strain. Naïve mice (n = 3) succumbed to infection rapidly, whereas vaccinated mice (n = 4) were resistant to challenge. This is one representative experiment and similar data were observed 1, 2, and 6 months postvaccination. WT, wild type; KO, knockout.

While these results suggest that IL-15 is not necessary for the development or maintenance of protective memory for T. gondii, they contrast with recent studies in which administration of soluble IL-15Rα was shown to enhance susceptibility to secondary challenge (13). In those studies, wild-type mice were first orally infected with a low dose of the 76K strain of T. gondii (similar to the ME49 strain) and 1 month later challenged with a high dose of the same strain in the presence of soluble IL-15Rα (13). When the same vaccination strategy was adopted in IL-15^−/− mice (low-dose ME49 followed by high-dose ME49 1 month later), these mice were resistant to challenge (data not shown). This discrepancy raises the concern that other cytokines in the IL-15^−/− mice may compensate for the loss of function of that gene, an effect that may not be observed in wild-type mice transiently treated with IL-15Rα. Alternatively, given the promiscuous use of cytokine receptors between members of the IL-2 family of cytokines (IL-2, IL-15, and IL-21) and the overlap in biological functions of these proteins, such as activation of T cells (9, 18), maturation of NK cells (2), and enhanced production of IFN-γ by NK or T cells (15), the use of soluble IL-15Rα may blockade multiple pathways in vivo that are involved in memory responses. Nevertheless, the studies presented here demonstrate that IL-15^−/− mice can develop and maintain protective T-cell responses to T. gondii infection. Similarly, infection of IL-15^−/− or IL-15Rα^−/− mice with lymphocytic choriomeningitis virus resulted in no defects in the primary immune response or in the development of memory CD8⁺ T cells, though the lymphocytic choriomeningitis virus model did reveal a defect in the maintenance of CD8 memory pools over time (1, 17). The data presented here indicate a limited role for IL-15 in the development and maintenance of NK and CD8⁺ T-cell responses required for resistance to T. gondii and are more consistent with studies in which the prominent role of IL-15 is in the development and homeostasis of these immune cells.